1. Got a question or need help troubleshooting? Post to the troubleshooting forum or Search the forums!

Cura 3.0

Discussion in 'Software' started by GriZag, Nov 17, 2017.

  1. GriZag

    GriZag New Member

    Joined:
    Nov 17, 2017
    Messages:
    8
    Likes Received:
    0
    I've been using the newest version of Cura from Ultimaker because it gives the option of a gradual infill. One of the biggest benefits I've noticed is the remarkable decrease in print times. For example, Cura for Robo estimated a time of 1 day, 19 hours for a print, and the same item with the same settings in Cura 3.0 was 17 hours and 27 minutes.

    I'm not super knowledgeable about how slicers go about identifying print times, but I printed in both settings and the times were very close to estimates and the quality of the prints were the same. Just thought I would share the info with the community!
     
  2. mark tomlinson

    mark tomlinson ༼ つ ◕_ ◕ ༽つ
    Staff Member

    Joined:
    Feb 21, 2013
    Messages:
    23,912
    Likes Received:
    7,338
    It is a guess. A scientific guess, but a guess. because to actually calculate it accurately would require a lot of calculus and a lot of thought :)

    Changes in acceleration and the like are not going to be exact and a lot of acceleration changes (a lot of directional changes will do that) would throw off the estimate.

    How they each do it is proprietary and some do a better job at guessing than others (and what you have your acceleration settings doing during the print matter).
     
  3. Day Vid

    Day Vid New Member

    Joined:
    Oct 27, 2017
    Messages:
    20
    Likes Received:
    3
    The thing I've noticed is that on the R2 itself I get a crazy estimate. For example, Cura may say 7 hours. When I load it it and print it the R2 console will estimate 30 hours. Then it actually takes closer to the 7 hours that Cura originally estimated. Is this normal?
     
  4. mark tomlinson

    mark tomlinson ༼ つ ◕_ ◕ ༽つ
    Staff Member

    Joined:
    Feb 21, 2013
    Messages:
    23,912
    Likes Received:
    7,338
    Yes, like I said it is always a guess by whichever software tells you the time.
    Sometimes the guessing is smarter. Sometimes not.
     
  5. John Wohlfeil

    John Wohlfeil Member

    Joined:
    Oct 13, 2017
    Messages:
    38
    Likes Received:
    15
    Ive noticed the same thing. I think its just like uploading or downloading files from the internet. It's trying to calculate a time on the fly and doing its best guess.

    My experience has been, the times estimated in Cure are close to the actual print time, but usually a little bit longer. I'm guessing that Cura isn't taking into account the warm up time and the bed leveling time, and that is where the little bit of extra time comes from. Otherwise, I think the Cura estimates have been pretty good.
     
  6. Geof

    Geof Volunteer Moderator
    Staff Member

    Joined:
    Nov 9, 2015
    Messages:
    6,757
    Likes Received:
    2,339
    All the estimates are hit or miss. The only the machine I have that is spot on is the Zortrax due to closed ecosystem. Their material, their slicer, their printers = dead on. All others are +/-. Should see a large format estimate be off, its not off 7 or so hours...more like 20 ...or 30... or more :D lol. Hit print and wait for it to finish ;) the bright side is that thing may be running its butt off, but we can all sit back with a cold one :D
     
  7. mark tomlinson

    mark tomlinson ༼ つ ◕_ ◕ ༽つ
    Staff Member

    Joined:
    Feb 21, 2013
    Messages:
    23,912
    Likes Received:
    7,338
    Totally model dependent (and on how you set various slicer options) on how close the "guesstimate" will be :) The one number you can take to the bank is the amount of filament it says it will use, that will be accurate. The print time? Not so much in most cases.
     
    Geof likes this.
  8. Kilrah

    Kilrah Well-Known Member

    Joined:
    Apr 18, 2017
    Messages:
    498
    Likes Received:
    332
    Yep. Octoprint uses several different estimation algorithms that evolve during print. At the start it kinda looks at how fast it builds up layers compared to total file layer count, but since the first layers are usually way longer to print due to being 100% infill and usually with slowdown for the first one it's VERY far off initially.
    Wait until you've done a dozen "normal" layers or so and it will have adjusted itself to be pretty accurate.
     
  9. supercazzola

    supercazzola Active Member

    Joined:
    Jun 1, 2017
    Messages:
    424
    Likes Received:
    111
  10. ytilotia

    ytilotia New Member

    Joined:
    Dec 14, 2018
    Messages:
    18
    Likes Received:
    4
    Question 1.
    I recently purchased an R2 (I had the original R1 from KickStarter).
    Can I assume that the Cura 2.5 version recommended by Robo can be upgraded to the latest (3.5) and as long as I install the R2 profile, I can use the new version?

    Question 2 -
    I'm not sure if I understand "Gradual Infill". Could someone either explain it or point me to some explanation?

    Thanks,
     
  11. mark tomlinson

    mark tomlinson ༼ つ ◕_ ◕ ༽つ
    Staff Member

    Joined:
    Feb 21, 2013
    Messages:
    23,912
    Likes Received:
    7,338
    Nice upgrade, I still have two of those working :)

    If you can apply the profile then sure it should work. (not speaking from experience as I don't use Cura).

    @WheresWaldo or @Geof might have better experience there.
     
  12. WheresWaldo

    WheresWaldo Volunteer ( ͠° ͟ʖ ͡°)
    Staff Member

    Joined:
    Feb 18, 2015
    Messages:
    5,905
    Likes Received:
    3,593
    Rarely used Cura, just looked at it a long while ago, been focusing my attention on Slic3r Prusa Edition at the moment. In some respects a better slicer than Simplify3D and of course the price is better too.
     
  13. ytilotia

    ytilotia New Member

    Joined:
    Dec 14, 2018
    Messages:
    18
    Likes Received:
    4
    Thanks for the Question 1 info. In my experience, the guys that are very active in an endeavor (like 3d printing) tend to gravitate to a software tool because it is "better" and not just "the one I started out with. While in my latest "learning" stage, I'd like to know what slicer is most popular without getting into a competition.
    So, if Cura isn't the "do all, be all, bellwether" for model prep, is Slic3r, SimplifyeD, or another more popular with the seasoned modelers?
    Opinion requested.

    Anyone what to take a stap at answering my question 2?
    Thanks,
     
  14. WheresWaldo

    WheresWaldo Volunteer ( ͠° ͟ʖ ͡°)
    Staff Member

    Joined:
    Feb 18, 2015
    Messages:
    5,905
    Likes Received:
    3,593
    Most people will gravitate to Simplify3D, for a long time it was way out in front of all the free slicers. There are a few quirks, but also a lot to like. It was the first slicer with really usable support generation.

    The problem with finding the "best" one by yourself is that each has its own learning curve and while one might be near perfect out of the box, another might require a bit of tweaking. As the technology stands today, I don't think there is a single "be all, end all" slicer out there. I also don't think there is a very bad one either (except maybe Matter Control [cumbersome] and Kiss Slicer [way to slow]).
     
  15. mark tomlinson

    mark tomlinson ༼ つ ◕_ ◕ ༽つ
    Staff Member

    Joined:
    Feb 21, 2013
    Messages:
    23,912
    Likes Received:
    7,338
    Never used it (specific to Cura/Ultimaker it appears). Some info here: https://community.ultimaker.com/topic/22778-gradual-infill-in-custom-setup/
     
  16. ytilotia

    ytilotia New Member

    Joined:
    Dec 14, 2018
    Messages:
    18
    Likes Received:
    4

    Thanks for the reply.
    As a veteran engineer/scientist for IBM research for over 30 years, I am used to tweaking, learning curves, and "the bleeding edge" in new technologies (I spent many of those years (1974 - 2005) developing Smart Homes and Premises...concepts, hardware, software and architecture.....long before they were vogue).
    My last 13 years (2005 - 2018) were spent in the CNC plasma cutting biz developing specialized electronics for controlling the z-axis while cutting steel plate at 60,000 degrees.....i.e Torch Height Controller
    That said, I very much enjoy the learning curves and opinions of "veteran makers" who have a vested insight on certain apps such as Cura, Simplify3D, Slic3r, etc. as well as 3D modelling app such as Fusion 360, etc.

    I welcome the experienced opinion.
    Thank you.
     
  17. ytilotia

    ytilotia New Member

    Joined:
    Dec 14, 2018
    Messages:
    18
    Likes Received:
    4
    Since Simplify3D cost $ instead of being free, can anyone who uses it tell me why they think its worth the cost? I don't mind paying for something if there is significant benefit in using it.
     
  18. mark tomlinson

    mark tomlinson ༼ つ ◕_ ◕ ༽つ
    Staff Member

    Joined:
    Feb 21, 2013
    Messages:
    23,912
    Likes Received:
    7,338
    Read through the threads :)

    Lots of previous discussion on it. Small sample here:

    http://community.robo3d.com/index.p...toprint-gcode-direct-to-octoprint-setup.8217/
    http://community.robo3d.com/index.php?threads/simplify-3d.1389/
    http://community.robo3d.com/index.php?threads/simplify-3d-profile-for-robo-r2.18639/
    http://community.robo3d.com/index.php?threads/simplify3d.21159/
    http://community.robo3d.com/index.php?threads/simplify3d-settings.20703/
    http://community.robo3d.com/index.php?threads/wow-simplify3d-and-r1-works-well.20295/


    For myself this is all I use and I have been using it since 2.0 (they are at 4.1.1 I think now)
    One-time, lifetime license w/support and so much cheaper than the CNC software...

    I usually tell folks to try the free ones until those let them down.
     
  19. WheresWaldo

    WheresWaldo Volunteer ( ͠° ͟ʖ ͡°)
    Staff Member

    Joined:
    Feb 18, 2015
    Messages:
    5,905
    Likes Received:
    3,593
    Right now the only real alternative to S3D is Slic3r, and specifically the Prusa Edition of Slic3r. Josef Prusa is pumping a lot of resources into refining Slic3r and it is on par and in some cases more intuitive than S3D. CNC people tell me that S3D is more akin to how a CNC operator thinks, but I cannot confirm that. I am also a long time S3D user but I am working with Slic3r now. What I have observed it that the filament profiles need a lot less tweaking than S3D, the "cubic" infill is stronger at lower densities than anything that S3D does. Still lacking is the support generation, in S3D it just works, no tweaking necessary. Even after dozens of prints I have to adjust support settings in Slic3r almost every time. I just haven't hit the right combination yet. Its close but not 100% there yet. Also the organization of S3D is just plain dumb in spots, stuff you think should be filament specific is global and some global stuff is specific to some other dependency, it is just weird. S3D since version 2 has an issue with thin wall sections and fill, it is at 4.1.1 and the issue is still there. S3D will slice complex models faster than Slic3r, but only marginally so. Of course I am using an 8 core 16 thread 32Gb current gen PC so we are only talking seconds, but there are still people using well past their useful life laptops and PCs and on those the speed difference will be more noticeable.
     
  20. ytilotia

    ytilotia New Member

    Joined:
    Dec 14, 2018
    Messages:
    18
    Likes Received:
    4
    Thanks for the input about the two.
    For the last 20 years I have been deep into CNC machining/creation endeavors including plasma, mill (Bridgeport and Sherline and everything in between), routers, lasers, and lathes (from Sherline to Clausing and several in between). Most have been powered by Mach3 , as it was very versatile.

    To me, the 3D priniting is just CNC that is additive instead of "subtractive"....though that may be a little simplified.
    In 3D printing, there needs to be the "adaptive" features as the material changes much the same as CNC Plasma cutting requires adaptive Torch Height Controllers because the heat affected zone changes and the torch height has to adapt to metal warpage, changing thickness (diamond plate), etc. Designing THC electronics was my business for over 10 years. Needless to say, I am very familiar with G-code.

    I very much want to get into the "knickers" of the slicers available and see which ones are a benefit for the things I want to do.
    If the Robo team isn't going to produce the second extruder option, I'd like to look at it a bit and see if I can make it happen. Otherwise, I'll purchase another brand of 3D printer that DOES have the extra extruder and keep both the R2 and the multi-extruder one for different build capabilities.

    BTW, has Robo released their firmware source or is what they use available somewhere so I can see the G-code commands they honor?

    Thanks again for the info. I look forward to contributing soon to the group.
     

Share This Page